Data cleaning
AuraScore 83/100

High-Frequency Execution Log Discrepancy Remediation Matrix

Standardize fragmented trade execution feeds and resolve microsecond timestamp drifts across trading venues.

Use this template when post-trade execution logs from disparate electronic communication networks exhibit dropped timestamps, phantom fills, or format mismatches. It generates a comprehensive data cleaning and reconciliation matrix for algorithmic trade compliance.

Template

Role: Principal Quantitative Data Architect with twenty years specializing in low-latency electronic trading infrastructure and post-trade tick reconciliation.

Context

  • Primary trading desk: {{trading_desk}}
  • Raw ingestion feeds: {{raw_feed_sources}}
  • Suspected anomaly types: {{suspected_anomaly_types}}
  • Matching tolerance window: {{matching_tolerance_window}}
  • Downstream surveillance platform: {{downstream_surveillance_system}}
  • Applicable regulatory reporting standard: {{regulatory_standard}}

Task

Construct an exhaustive execution feed sanitization matrix that diagnoses malformed, drifted, or duplicate order lifecycle events across {{raw_feed_sources}} and prescribes deterministic imputation, deduplication, and normalization rules compliant with {{regulatory_standard}}.

Method

  1. Profile timestamp precision disparities across {{raw_feed_sources}}, standardizing UTC offsets and sub-millisecond drift.
  2. Isolate orphan execution fills lacking corresponding parent order identifiers within the {{matching_tolerance_window}}.
  3. Identify synthetic duplicate trade messages generated during venue failover events across {{trading_desk}} feeds.
  4. Define deterministic tie-breaking logic for conflicting execution prices on fragmented book feeds.
  5. Map null, truncated, or non-standard MIC codes and broker IDs to ISO 10383 identifiers.
  6. Formulate fallback imputation strategies for missing counterparty identifiers without corrupting regulatory audit trails.
  7. Establish threshold parameters where unresolvable records must be quarantined prior to {{downstream_surveillance_system}} ingestion.

Constraints

  • MUST maintain complete raw-to-transformed lineage for every corrected execution attribute.
  • MUST NOT drop unparseable rows without assigning a standardized error quarantine code.
  • Cleaning rules must preserve deterministic reproducibility for retrospective market abuse audits.
  • Matrix actions must clearly distinguish between automated transformations and required quant oversight.

Output format

A comprehensive Markdown document containing: 1) Feed Ingestion Architecture Summary (max 150 words); 2) Execution Cleaning Matrix (a structured table containing Columns: Feed Source, Field Name, Anomaly Pattern, Root Cause, Transformation Rule, Preservation Logic, and Quarantine Trigger); 3) Post-Sanitization Validation Checks (numbered list of exactly 5 statistical assertions).

Self-review

  • Ensure every anomaly listed in {{suspected_anomaly_types}} has a dedicated row in the matrix.
  • Verify that timestamp reconciliation logic respects {{matching_tolerance_window}} constraints.
  • Confirm that no destructive cleaning rules violate {{regulatory_standard}} audit mandates.
AuraScore breakdown
83/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering12/12 · Strong

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness5/5 · Strong

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

data-analytics
data-cleaning
financial-services
data cleaning
trade reconciliation
fintech